Hasselblad X1D II

[anfat]

I would like to ask Hasselblad X1D II 50C owners what exposure metering method they use and if it is possible to perform ETTR, since there is no histogram in live view, thanks

 

[Godfrey]

It's the same with the 907x ... no live histogram. I generally use aperture priority automatic or manual exposure, and normally have the metering pattern set to center-weighted.

I never needed a live view histogram to use center-weighted aperture priority with generations of film cameras, and have always considered ETTR as the right technique for negative films; the digital cameras makes it easy. With digital, you ust make an exposure and see if you're blowing out any of the important highs when you have a contrasty light situation. If you are, back down the exposure a little bit until the highs are not blown out. Once you know that correction for a typical scene, just use it for all the scenes in similar light.

G

 

[anfat]

Thank you...in the days of film I used a spot meter and placed the lights as high as possible ...zone VII or VIII

 

[Godfrey]

If that worked well for you (and it does for me in the more/most difficult lighting situations), there's no reason to not just keep on doing it. The issue of exposure always comes down to what you choose to discard if you're working in contrasty light beyond the dynamic range of the sensor to capture. The X1D II and 907x sensors have enormous dynamic range so do whatever works to keep from hitting highs saturation, you can get the rest out easily during raw conversion processing.

G

 

[SrMphoto]

Here is a method that works without live histogram or highlight warnings. I have not mastered it yet , and prefer the live highlight warnings instead.

https://www.fastrawviewer.com/blog/spot-meter-exposure

One of the ways of getting good exposure is metering while using the in-camera spotmeter on the lightest part of the scene that needs to maintain full detail (white clouds, snow, etc.) and applying the appropriate compensation to the exposure recommended by the spotmeter.

 

[OleBe]

To give you one example of how I do it. Outside sky with white clouds. I measure for the brightness of the whitest cloud a spot metering of +2,6. That way the camera might think already that a certain amount of the whites are clipping and gives you a small amount of clipping warning in the image preview, but in reality they are not (you have headroom in the same example to nearly +4 spot metering). Hasselblad told me that this is due to the jpg conversation in camera and on their list to get improved as this throws away information, which you easily could get by exposing to full sensor flooding.

 

[SrMphoto]

To give you one example of how I do it. Outside sky with white clouds. I measure for the brightness of the whitest cloud a spot metering of +2,6. That way the camera might think already that a certain amount of the whites are clipping and gives you a small amount of clipping warning in the image preview, but in reality they are not (you have headroom in the same example to nearly +4 spot metering). Hasselblad told me that this is due to the jpg conversation in camera and on their list to get improved as this throws away information, which you easily could get by exposing to full sensor flooding.

Almost all manufacturers provide only JPG based histograms and highlight warning, whereas we need raw-based histogram to expose optimally. This leads to strategies like adapting in-camera JPG settings or using UniWB. If Hasselblad could come up with raw based live histograms, that would be fantastic.

 

[Godfrey]

The problem with implementing a raw-based histogram is that a raw file is undisplayable until demosaic and gamma correction is done. So even working directly with the raw data, you must do demosaic and gamma correction according to SOME parametric settings for the defaults to obtain usable results... which is typically what they use the JPEG engine to do in-camera. It is probably going to be as good as anything you're going to get from a "raw" based histogram to use whatever the JPEG image settings you like as a basis for the analysis, with the understanding that the raw data will have a bit more overhead on highlight saturation and intermediary values at the shadow end.

(I know that the histogram in tools like LR generally run a demosaic and gamma correction according to the baseline raw conversion settings established by the camera calibration profile and pull the values out to make the histogram on that basis. This is very similar to 'basing the histogram on the JPEGs' that you see happening in-camera.)

G

 

[SrMphoto]

The problem with implementing a raw-based histogram is that a raw file is undisplayable until demosaic and gamma correction is done. So even working directly with the raw data, you must do demosaic and gamma correction according to SOME parametric settings for the defaults to obtain usable results... which is typically what they use the JPEG engine to do in-camera. It is probably going to be as good as anything you're going to get from a "raw" based histogram to use whatever the JPEG image settings you like as a basis for the analysis, with the understanding that the raw data will have a bit more overhead on highlight saturation and intermediary values at the shadow end.

(I know that the histogram in tools like LR generally run a demosaic and gamma correction according to the baseline raw conversion settings established by the camera calibration profile and pull the values out to make the histogram on that basis. This is very similar to 'basing the histogram on the JPEGs' that you see happening in-camera.)

G

I think that we are most interested in the clipping/sensel-saturation information in order to adjust the exposure to avoid clipping relevant highlights. That information is independent of demosaicing and gamma correction and is not extractable from JPG histograms/blinkies. Note that histograms in LrC and especially Photoshop are not that accurate in that regard. Currently, the only reliable method to check for clipping is RawDigger software.

 

[PabloR]

With this camera you are looking the sensor response, live !!

Why do you need anything more ?

Don't have to meter zones anymore. today metering is as fast that it will do not have any complication anymore, never !

So, if the camera finally is not an obstacle, if there will not be technical difficulties, it is time to focus on the important things.

this camera is stunning !!

 

[SrMphoto]

With this camera you are looking the sensor response, live !!

Why do you need anything more ?

Don't have to meter zones anymore. today metering is as fast that it will do not have any complication anymore, never !

So, if the camera finally is not an obstacle, if there will not be technical difficulties, it is time to focus on the important things.

this camera is stunning !!

Mirrorless cameras are not 100% WYSIWYG. That is why we have blinkies and histograms in the review.
Are you asking why we should expose optimally (aka saturated sensels without clipping relevant highlights)? That question is best discussed in a different thread .

 

[msstudio]

In questionable circumstances I use my lightmeter that’s calibrated. And/or bracket. Mostly trying to keep the highlights from going out of range. And as a last resort, shoot, look at the curves and highlights and make a creative decision.

 

[Godfrey]

I think that we are most interested in the clipping/sensel-saturation information in order to adjust the exposure to avoid clipping relevant highlights. That information is independent of demosaicing and gamma correction and is not extractable from JPG histograms/blinkies. Note that histograms in LrC and especially Photoshop are not that accurate in that regard. Currently, the only reliable method to check for clipping is RawDigger software.

I've never used "Rawdigger" and do not have any need for it. I am able to make excellent exposures that are easy to render and meet my desires for a photograph without being obsessed over "clipping/sensel-saturation information" ... and without clipping highs or otherwise causing myself problems. I can easily tell when I've burned highlights. It's not hard to do. My goal is to make photographs ... I just don't see the necessity of being obsessed with technicalities.

G

 

[FloatingLens]

The problem with implementing a raw-based histogram is that a raw file is undisplayable until demosaic and gamma correction is done.

That may be true, but I don't get why it prevents the software from displaying weighted RGB distribution straight from the individual raw channels just for judging if any of the channels is near the saturation point. Could anybody elaborate?

As long as the histogram data is not based on processed JPEG data, I have no problem if the image preview beneath is rendered based on JPEG. Then again, my understanding of the demosaic process as contributor to compute a full histogram maybe incomplete.

 

[Godfrey]

Processing power: To do that requires you demosaic and gamma correct the raw data first and then run the numbers on the channel oriented data. To do this at full resolution takes a lot of memory and processing power, and then evaluating a histogram on full resolution data again takes a good deal of memory and processing power. It's not practical for the low powered processors and limited memory in a camera device.

The camera does this transformation at low resolution, quickly, when generating an exposure's JPEG preview which is constrained to 8bit RGB channel data. So the histogram data is based on the data in the JPEG preview, not on the actual raw data itself.

G